A typical Wireless Local Area Network (WLAN) deployment consists of a collection of Access Points (APs) spread over a geographical area. The placement of each Access Point within the geographical area is determined via a radio frequency (RF) Planning tool. The RF planning tool is a wireless deployment modeling tool that helps the administrator design an efficient network that optimizes coverage and capacity.
Coverage relates to the geographical footprint within the system that has sufficient Radio Frequency (RF) signal strength to provide for a call/data session. Capacity relates to the capability of the system to sustain a given number of subscribers. Capacity and coverage are two important factors in the WLAN deployment. The network administrator needs to carefully plan the deployment keeping in mind the typical number of users, the network load and the expected levels of performance. In a typical network, the network administrator deploys the APs based on a set of use-cases that provide him or her estimates for (a the average number of users that would need to be serviced by the access points; and (b) the average traffic throughput that would need to be serviced by the access points.
Moreover, when the WLAN technology first came into existence, the WLAN was thought of as an overlay network that could be used as an extension of the wired network without promising any kind of reliability, security and performance guarantees. With the maturity of WLAN technology over time, currently, the WLAN is capable of providing similar levels of reliability, security and performance guarantees as those of a wired network.
In addition, with more and more enterprises adopting the “Bring Your Own Device (BYOD)” policy, the typical user has multiple wireless devices connected to the network. For example, a user might have a laptop, a smartphone, a tablet, an e-book reader, etc., all of which are connected to the network. Further, the devices are all capable of transmitting and receiving various categories of traffic, including but are not limited to, voice, video, bit-torrent, general data traffic, etc. With this kind of surge in device numbers and traffic amount, many network deployments would be placed under severe stress.
Call Admission Control (CAC) generally refers to a mechanism that protects voice traffic from the negative effects of other voice traffic and to keep excess voice traffic off the network. For example, a network administrator may limit the maximum number of voice users in a particular network. As another example, the network administrator may a given AP may not admit additional voice calls if the AP is currently handling more than a threshold amount of voice traffic. Nevertheless, the CAC mechanisms do not solve the problem if a non-voice client starts a large amount of upload/download after a call has been admitted. There is no current mechanism for limiting the total number of clients once the maximum number of voice clients has been reached. Also, note that, CAC mechanism is typically used in a wired network where it is easy to predict and allocate bandwidth for wired clients. With interferences in a wireless network, an AP's allocation of bandwidth is typically only good for the instant the allocation was made because the network condition (e.g., interference levels) may change over time.
Also, some client devices have a preference for being associated with the same AP after it enters a voice call. For a client device with such a preference, if a voice call is initiated when the client device is located in a good coverage area, but the client device subsequently moves to a poor coverage area, the client device will not associate with a new AP. As a result, the client device will suffer from poor call quality after it moves to the new physical location.
In addition, conventional ClientMatch™ mechanism performs client steering when the signal strength, e.g., Signal-to-Noise Ratio (SNR), is below a threshold value. Thus, if the client continues to have good SNR, the client will keep being associated with the same AP, even though the AP may have been loaded with other clients doing heavy data transfer and other adjacent APs are lightly loaded.
Moreover, ClientMatch™ mechanism steers the clients to adjacent APs irrespective of the nature of sessions currently in transit for that client. Therefore, if the sessions in transit include latency sensitive traffic, for example, voice and/or video traffic, these sessions may experience undesirable latency or jitter if the client is steered to some other AP. Hence, steering clients based on SNR value alone may not be applicable for all use case scenarios.